ccna practical cisco routers phần 2 ppsx

3 1
PART I
Understanding Network Arch itect ures CHAPTER 1
AppleTalk uses a special dynamic addressing system to determine the
address of the nodes on the network. When a Macintosh is powered
up on the network, the computer generates a random address and
broadcasts it out onto the network. This random address becomes its
network address (if another Macintosh isn’t already using that
address; if so, the newly powered on Mac will continue to generate
random addresses until it finds one that is unused).
AppleTalk is similar to Ethernet in that it is a passive network archi-
tecture. AppleTalk uses Carrier Sense multiple access with collision
detection—CSMA/CA. Basically the computers sit on the network
and listen to determine whether the wire is clear. After making sure
the network is clear, the computer will send a packet onto the net-
work letting all the other computers know that it intends to transmit
data. The computer then sends out its data.
The fact that a computer that intends to send data out onto the net-
work notifies the other network nodes as to its intentions greatly
reduces the number of collisions on a CSMA/CA network (especially
when compared to Ethernet).
These announcement packets, however, do have a tendency to slow
down the network and Macintosh networks only have a transmission
speed of 230.4 Kbps. The fact that the hardware and software
needed to network a group of Macintosh computers comes with each
Macintosh (other than the LocalTalk cable) makes it an easy and
inexpensive way to network several workstations to share a printer or
files.

The OSI Model and Network
P r o t o c o l s

OSI—The Theoretical Networking
•
Protocol Stack
The OSI Layers
•
The Data Link Sublayers
•
Real-World Network Protocols
•
2
c h a p t e r
3 4
OSI—The Theoretical Networking Protocol
Stack
Conceptual models are something that you run into no matter what
discipline you tackle. Art embraces color and design theories; physics
embraces nearly every theoretical model that Einstein scrawled on a
napkin. Computer networking is no different and it also uses a con-
ceptual model or framework that allows us to discuss a complex
chain of events—data movement on a network.
In the late 1970s the International Standards Organization (ISO)
began to develop a conceptual model for networking called the Open
Systems Interconnection Reference Model. Networking folk more com-
monly refer to it as the OSI model (and I’m sure a number of them
have forgotten what the OSI stands for). In 1984, the model became
the international standard for network communications, providing a
conceptual framework that helps explain how data gets from one
place to another on a network.
The OSI model describes network communication as a series of
seven layers that operate in a stack; each layer is responsible for a dif-

ferent part of the overall process of moving data. This framework of
a layered stack, while conceptual, can then be used to discuss and
understand actual protocol stacks that we see used for networking.
For example, TCP/IP and AppleTalk are two real-world network
protocol stacks; protocols that actually serve as layers in a protocol
suite like TCP/IP can then be discussed in terms of how they relate
to and serve at various levels of the OSI model’s stack.
SEE ALSO
➤ To learn more about several of the commonly used network protocol suites,see page 44.
The OSI model provides the model for a number of important
events that take place during network communication. It provides
basic rules of thumb for a number of different networking processes:
■ How data is translated into a format appropriate for your net-
work architecture. When you send an email or a file to another
computer, you are working with a certain application such as an
email client or an FTP client. The data you transmit using this
application must be placed in a more generic format if it is going
to move out onto the network and to the intended recipient.
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Model and Netwo rk Pro tocols
ISO seems to ring a bell
The International Standards
Organization (ISO) is
involved in developing sets
of rules and models for
everything from technical
standards for networking to
how companies do busi-
ness in the new global
market. You’ve probably

seen banners on busi-
nesses announcing that
they are ISO 9002 certified.
This means that they are in
compliance with the set of
rules and protocols that
have been developed by
the ISO for doing business
in the world marketplace.
Another common ISO certi-
fication—ISO 9660—
defines file systems for
CD-ROMs.
3 5
PART I
The OSI Layers CHAPTER 2
■ How PCs or other devices on the network establish communica-
tions. When you send data from your PC, there must be some
mechanism that supplies a communication channel between
sender and receiver. It’s not unlike picking up a telephone and
making a call.
■ How data is sent between devices and how sequencing and error
checking is handled. After a communications session has been
established between computers, there must be a set of rules that
controls how the data passes between them.
■ How logical addressing of packets is converted to the actual
physical addressing provided by the network. Computer net-
works use logical addressing schemes such as IP addresses.
There must be a conversion of these logical addresses to the
actual hardware addresses found on the NICs in the computers.

The OSI model provides the mechanisms and rules that make the
handling of the issues discussed in the bulleted list possible.
Understanding the various layers of the OSI model not only provides
insight into actual network protocol suites, but it also provides you
with a conceptual framework that can be used to better understand
complex networking devices like switches, bridges and routers.
(Much of this book is devoted to a discussion of routers and routing.)
The OSI Layers
The layers of the OSI model explain the process of moving data on a
network. As a computer user, the only two layers of the model that
you actually interface with are the first layer—the Physical layer—
and the last layer—the Applications layer.
■ The Physical layer constitutes the physical aspects of the network
(the network cabling, hubs, and so on). You’ve probably inter-
faced with the physical layer at least once, when you tripped over
a poorly situated cable.
■ The Application layer provides the interface that you use on your
computer to send email or place a file on the network.
Obviously, this would be a very short chapter if we only discussed
these two layers, but you will find each and every layer of the OSI
model plays an important part in the networking of information.
So, what’s a protocol
stack?
Protocol stacks orsuites(or
layers) are a group of small
protocols that work
together to accomplish the
movement of data from one
node on a network to
another. Protocol stacks are

not unlike relay-race run-
ners, although packets of
data rather than a baton
are handed off to each sub -
sequent protocol until the
packets of data are in a
form (a single bit stream)
that canbe placed on the
network medium.
The ISO/OSI protocol
stack exists!
While network protocol
stacks like NetWare’s
IPX/SPX and TCP/IP are
something with which most
network administrators are
quite familiar, there is
actually a real protocol
suite based on the OSI
model; it’s called the OSI
protocol stack.
Unfortunately, it is not
embraced by any of the
network operating systems
(such as Novell NetWare or
Windows NT) with which
you will actuallywork.
3 6
Figure 2.1 provides a list of the OSI model layers from the top of the
stack to the bottom. An upside-down pyramid is also an apt

representation of the model because data is taken in a fairly complex
form and eventually converted to a simple bit stream that can be
placed on the network wire. You will notice that the layers are num-
bered, however, from top to bottom. For instance, in a discussion of
the Network layer, you may hear the layer described as Layer 3.
Whether you use the name or number is unimportant; you just need
to make sure that you understand the role of each layer in the overall
process of data communications.
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Mode l an d Network Protocols
FIGURE 2.1
The OSI model provides
a conceptual basis for
how data moves from a
sending computer to a
receiving computer.
A good way to remember the network layers from bottom to top is
the following mnemonic: Please Do Not Throw Sausage Pizza
Away. And (unfortunately, you may be thinking), you really do need
to remember the OSI model; it is important to any discussion of net-
working technology from the very simple to the very complex. Every
book or article you pick up on networking will make some reference
to the model.
Before we discuss each of the layers in the stack, it makes sense to get
a general idea of what takes place when data moves through the OSI
model. Let’s say that a user decides to send an email message to
another user on a network. The user sending the email will take
advantage of an email client or program (such as Outlook or Eudora)
that serves as the interface tool where the message is composed and
then sent. This user activity takes place at the Application layer.

3 7
PART I
The OSI Layers CHAPTER 2
After the data leaves the Application layer (the layer will affix an
Application layer header to the data packet) it moves down through
the other layers of the OSI stack. Each layer in turn does its part by
providing specific services related to the communication link that
must be established, or by formatting the data a particular way.
No matter what the function of a particular layer is, it adds header
information (the headers are represented as small boxes on Figure
2.2) to the data. (The Physical layer is hardware—a cable, for
instance—so it doesn’t add a header to the data.)
The data eventually reaches the Physical layer (the actual network
medium such as twisted pair cable and the hubs connecting the com-
puter) of the email sender’s computer and moves out onto the net-
work media and to its final destination—the intended recipient of the
email.
FIGURE 2.2
Data moves down
through the OSI stack of
the sending computer
and moves up through
the OSI stack on the
receiving computer.
Application layer
header
Presentation layer
header
Packet with full com-
plement of OSI layer

headers
Headers are removed
as the datamoves up
the OSI stack
The data is received at the Physical layer of the recipient’s computer
and moves back up through the OSI stack. As the data moves
through each layer, the appropriate header is stripped from the data.
When the data finally reaches the Application layer, the recipient
can use his or her email client to read the received message.
3 8
The following discussion of the OSI layers will discuss the layers in
the stack from top to bottom (Application layer to Physical layer).
The Application Layer
The Application layer provides the interface and services that sup-
port user applications. It is also responsible for general access to the
network.
This layer provides the tools that the user actually sees. It also pro-
vides network services related to these user applications such as mes-
sage handling, file transfer, and database queries. Each of these
services are supplied by the Application layer to the various applica-
tions available to the user. Examples of information exchange ser-
vices handled by the Application layer would include the World
Wide Web, email services (such as the Simple Mail Transfer
Protocol—more commonly referred to as SMTP—found in
TCP/IP), and special client/server database applications.
The Presentation Layer
The Presentation layer can be considered the translator of the OSI
model. This layer takes the packets (packet creation for the move-
ment of the data to the network actually begins in the Application
layer) from the Application layer and converts it into a generic for-

mat that can be read by all computers. For instance, data represented
by ASCII characters will be translated to an even more basic, generic
format.
The Presentation layer is also responsible for data encryption (if
required by the application used in the Application layer) and data
compression that will reduce the size of the data. The packet created
by the Presentation layer is pretty much the final form that the data
will take as it travels down through the rest of the OSI stack
(although there will be some additions to the packets by subsequent
layers and data may be broken into smaller packet sizes).
The Session Layer
The Session layer is responsible for setting up the communication
link or session between the sending and receiving computers. This
layer also manages the session that is set up between these nodes (see
Figure 2.3).
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Model and Netwo rk Pro tocols
Communications take
place between peer
layers
While data movesdown
through the protocol stack
on the sender’s computer
(such as an email message)
and eventually out onto the
wire and then up the proto-
col stack on the receiving
computer, communications
do take place between
complementary layers on

each computer. For exam-
ple, there is virtual commu-
nication between two
computers sending and
receiving data at the
Session layer. Which
makes sense because this
is the layer that controls
the communication
between the two comput-
ers over the network media
(which could be twisted
pair wire, fiber opticwire,
or other connective media).
3 9
PART I
The OSI Layers CHAPTER 2
After the session is set up between the participating nodes, the
Session layer is also responsible for placing checkpoints in the data
stream. This provides some fault tolerance to the communication
session. If a session fails and communication is lost between the
nodes, once the session is reestablished only the data after the most
recently received checkpoint will need to be resent. This negates the
need to tie up the network by resending all the packets involved in
the session.
Actual protocols that operate at the Session layer can provide two
different types of approaches to getting the data from sender to
receiver: connection-oriented communication and connectionless
communication.
Connection-oriented protocols that operate at the Session layer pro-

vide a session environment where communicating computers agree
upon parameters related to the creation of checkpoints in the data,
maintain a dialogue during data transfer, and then simultaneously
end the transfer session.
Connection-oriented protocols operate much like a telephone call:
You establish a session with the person you are calling. A direct con-
nection is maintained between you and the party on the other end of
the line. And when the discussion concludes both parties typically
agree to end the session.
Connectionless protocols operate more like the regular mail system.
They provide appropriate addressing for the packets that must be
sent and then the packets are sent off much like a letter dropped in
the mailbox. It is assumed that the addressing on the letter will get it
to its final destination, but no acknowledgment is required from the
computer that is the intended destination.
Users must run the
same protocol stack to
communicate
In the previous example of
an email message being
sent and received, it was
assumed that both the
sender and receiver of the
data involved were running
the same protocol stack
(the theoretical OSI stack)
on their client computers.
Very different computers
running very different oper-
ating systems can still

communicate if they
embrace a common net-
work protocol stack. This is
why a UNIX machine, an
Apple Macintosh, or a PC
running Windows all use
TCP/IP to communicate on
the Internet. A case where
two computers could not
communicate would be
where a computer running
TCP/IP is trying to commu-
nicate with a computer that
is only running IPX/SPX.
Both of these real-world
protocols use different
rules and data formats,
makingcommunication
impossible.
FIGURE 2.3
The Session layer pro-
vides the communication
link between the two
communicating
computers.
4 0
The Transport Layer
The Transport layer is responsible for the flow control of data
between the communicating nodes; data must not only be delivered
error-free but also in the proper sequence. The Transport layer is

also responsible for sizing the packets so that they are in a size
required by the lower layers of the protocol stack. This packet size is
dictated by the network architecture.
SEE ALSO
➤ For more about network architectures such as Ethernet and Token Ring,see page 25.
Communication also takes place between peer computers (the sender
and receiver); acknowledgements are received from the destination
node when an agreed upon number of data packets have been sent by
the sending node. For example, the sending node may send three
bursts of packets to the receiving node and then receive an acknowl-
edgement from the receiver. The sender can then send another three
bursts of data.
This communication at the Transport layer is also useful in cases
where the sending computer may flood the receiving computer with
data. The receiving node will take as much data as it can hold and
then send a “not ready” signal if additional data is sent. After the
receiving computer has processed the data and is able to receive
additional packets, it will supply the sending computer with a “go-
ahead” message.
The Network Layer
The Network layer addresses packets for delivery and is also respon-
sible for their delivery. Route determination takes place at this layer,
as does the actual switching of packets onto that route. Layer 3 is
where logical addresses (such as the IP address of a network com-
puter) are translated to physical addresses (the hardware address of
the NIC—Network Interface Card—on that particular computer).
Routers operate at the Network layer and use Layer 3 routing proto-
cols to determine the path for data packets.
How routes are determined and how routers convert logical
addresses to physical addresses are subjects that we will look at in

much more detail throughout this book.
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Model and Netwo rk Pro tocols
Application layer ser-
vices make user appli-
cations work over the
network
When a user working in a
particular application
(Excel, for example) decides
to save a worksheet file to
his or her home directory
on the network file server,
the Application layer of the
OSI model provides the
appropriate service that
allows the file to be moved
from the client machine to
the appropriate network
volume. This transaction is
transparent to the user.
Each layer performs
functions on outgoing
and incoming data
Remember that each layer
in the OSI model (or in an
actual network protocol
stack such as IPX/SPX or
TCP/IP) have responsibili-
ties related to outgoing and

incoming information.
When data is moving down
the stack on a sending
computer, the Presentation
layer converts information
from a particular applica-
tion to a generic format. On
the receiving computer the
Presentation layer would
take generic information
moving up the OSI stack
and convert it into a format
usable by the appropriate
Application layer program
on the receiving computer.
4 1
PART I
The OSI Layers CHAPTER 2
SEE ALSO
➤ Our discussion of the Network layer will be greatly expanded in later chapters. To begin an
exploration of how routers operate at the Network layer see page 77.
The Data-Link Layer
When the data packets reach the Data-Link layer, they are placed in
data frames defined by the network architecture embraced by your
network (such as Ethernet, Token Ring, and so on). The Data-Link
layer is responsible for data movement across the actual physical link
to the receiving node and so uniquely identifies each computer on
the network based on its hardware address that is encoded into the
NIC (Network Interface Card). Figure 2.4 shows the hardware
address for the network interface card used in a networked computer

running Windows 98.
Real-world protocols
use a combination of
connection-oriented and
connectionless commu-
nication
You will find that in net-
work protocol stacks—
such as TCP/IP and
IPX/SPX—both connection-
orientedand connection-
lesscommunication
strategies are used to
move data on the network.
Typically, more than one
protocol will operate at the
Sessionlayer to handle
these different
communication strategies.
FIGURE 2.4
Each node on the net-
work will have a unique
physical address.
Header information is added to each frame containing the sending
address and the destination address. The Data Link layer is also
responsible for making sure that the frames sent over the physical
link are received error-free. So, protocols operating at this layer will
add a Cyclical Redundancy check (CRC) as a trailer on each frame. The
CRC is basically a mathematical calculation that takes place on the
sending computer and then on the receiving computer. If the two

CRCs match up, the frame was received in total and its integrity was
maintained during transfer.
4 2
Again, as mentioned earlier, the frame type produced by the Data
Link layer will depend on the network architecture that your net-
work embraces, such as Ethernet, IBM Token Ring, or FDDI.
Figure 2.5 shows an Ethernet 802.2 frame. Table 2.2 lists and
describes each of the frame components. While you may not fully
understand all the parts of the frame shown, note that the makeup of
the frame is basically header information that describes the frame,
the actual data in the frame, and then Data-link layer information
(such as Destination Service Access Points and Service Access Points)
that not only define the Frame type (in this case Ethernet) but also
serve to help get the frame to the receiving computer. (For more
about the IEEE 802 specifications, see the “Ethernet Frame Trivia”
sidebar.)
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Model and Netwo rk Pro tocols
FIGURE 2.5
The Ethernet frame is
created at the Data Link
layer of the OSI model.
Table 2.2 Ethernet Frame Segments
Segment Purpose
Preamble Alternating bits (1s and Os) that announces that a frame has been sent
Destination The destination address
Source The source address
Length Specifies the number of bytes of data in the frame
DSAP Destination Service Access Point—this tells the receiving network
card where to place the frame in buffer memory

SSAP Provides the Service Access Point information for the frame (Service
Access points are discussed in the “Data-Link section later in this
chapter)
CTRL A Logical Link control field (Logical Link control is discussed in the
“Data-Link Sublayers” section later in this chapter).
Data This part of the frame holds the actual data being sent
FCS Frame Check Sequence field contains the CRC value for the frame
4 3
PART I
The Data-Link Sublay ers CHAPTER 2
The Data Link layer also controls how computers access the physical
network connections. This aspect of Layer 2 will be discussed more
fully in the “Data Link Sublayers” section that follows this discussion
of the OSI layers.
The Physical Layer
At the Physical layer the frames passed down from the Data Link
layer are converted into a single bit stream that can then be sent out
onto the network media. The Physical layer also defines the actual
physical aspects of how the cabling is hooked to the computer’s NIC.
On a computer that is receiving data, the Physical layer receives the
bit stream (information consisting of 1s and 0s).
SEE ALSO
➤ To learn more about the commonly used network media and cable types, see page 17.
The Data-Link Sublayers
Before we end our discussion of the OSI networking model, we need
to back track a little and discuss additional specifications that were
developed for the Data Link layer of the OSI model by the IEEE.
The IEEE 802 specifications divided the Data Link layer into two
sublayers: Logical Link Control (LLC) and Media Access Control
(MAC).

The Logical Link Control sublayer establishes and maintains the link
between the sending and receiving computer as data moves across
the network’s physical media. The LLC sublayer also provides
Service Access Points (SAPs), which are reference points that other
computers sending information can refer to and use to communicate
with the upper layers of the OSI stack on a particular receiving node.
The IEEE specification that defines the LLC layer is 802.2 (see
IEEE specifications sidebar for more information on the categories).
Finding MAC addresses
on Windows computers
To find the address of a
network card running on a
Windows 95/98 computer,
click the Start menu, and
then click Run. In the Run
dialog box, type winipcfg,
and then click OK. The IP
Configuration dialog box
will appear for the com-
puter and provide the
address for the Network
card. On a Windows NT
computer, right-click on the
Network Neighborhood
icon and then select the
Adapters tab on the
Network dialog box. Select
your network adapter and
then click the Properties
button. The MAC address

of the NIC should be
provided.
4 4
The Media Access Control sublayer determines how computers
communicate on the network and how and when a computer can
actually access the network media and send data. The 802 specifica-
tions actually break the MAC sublayer down into a list of categories
(ways of accessing the network media) that directly relate to specific
network architectures such as Ethernet and Token Ring (see
Figure 2.6).
SEE ALSO
➤ For more information on some of the common network architectures like Ethernet and Token
Ring, see page 25.
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Model and Netwo rk P rotocols
FIGURE 2.6
The Data Link Layer con-
sists of two sublayers:
theLLC and the MAC.
Real-World Network Protocols
Now that we’ve taken a look at the theoretical model for how data
moves from one computer to another on a network, as seen in the
different layers of the OSI model, we can take a look at some of the
most commonly used network protocol stacks and map their differ-
ent layers to the OSI model. This will provide you with a good
understanding of how these real-world protocol stacks operate and
provide data transport on the network.
You will also see which protocols in a particular protocol stack are
involved at the Network layer of the OSI model. These protocols
will become important as we discuss the routing of packets on an

Internetwork (something that we will do for much of the book).
4 5
PART I
R e a l - W orl d Ne twork Proto cols CHAPTER 2
NetBEUI
NetBEUI (NetBIOS Extended User Interface) is a simple and fast net-
work protocol that was designed to be used with Microsoft’s and
IBM’s NetBIOS (Network Basic Input Output System) protocol in
small networks. NetBEUI operates at the Transport and Network
layers of the OSI model.
Because NetBEUI provides only the services needed at the Transport
and Network layers of the OSI stack, it needs NetBIOS, which oper-
ates at the Session layer of the OSI stack, and is responsible for set-
ting up the communication session between two computers on the
network. Two other networking components found in Microsoft net-
works are the Redirector and the Server Message Block. The
Redirector operates at the Application layer and makes a client com-
puter perceive all the network resources as if they were local. Server
Message Block (SMB) provides peer-to-peer communication
between the Redirectors on client and network server machines. The
Server Message Block operates at the Presentation layer of the OSI
model.
While an excellent transport protocol with very low overhead,
NetBEUI is not a routable protocol, so it cannot be used on
Internetworks where routing takes place. This means that while you
should remember NetBEUI as a network protocol possibility for
small, simple networks, it is not an option for larger networks that
make use of routers (and so this is the last time you will hear about
NetBEUI in this book).
TCP/IP

Often referred to as the “protocol of low bid” (see the TCP/IP
Trivia sidebar for more information on TCP/IP’s interesting gene-
sis), TCP/IP has become the de-facto standard for enterprise net-
working. TCP/IP networks are highly scalable, so TCP/IP can be
used for small or large networks.
A word about hardware
addresses
NIC hardware addresses
are also called MAC
Addresses. MAC stands for
Media Access Control and
it is one of the sublayers of
the Data-Link layer (the
MAC sublayer will be dis-
cussed in the “Data-Link
Sublayers” section later in
this chapter). Hardware
addresses are burned onto
ROM chips on network
interface cards, giving each
of them a unique address.
The addressing scheme
was developed by the
Institute for Electrical and
Electronic Engineers (IEEE).
The actual address takes
the form of a 48-bit
address that is written in
hexadecimal format. An
example of a MAC address

is 00-00-B3-83-B3-3F.
Ethernet frame trivia
The Ethernet frame used by
early versions of Novell
NetWare (NetWare 2.x and
3.x) was created before the
IEEE specifications were
completed. This means that
The Ethernet 802.3 frame
type is actually not to
specifications as outlined
by the IEEE. New versions
of NetWare and other
Ethernet network operating
systems now use the 802.2
Ethernet frame, which is
completely compliant with
the IEEE specifications(the
IEEE specifications are
listed later in this chapter).
4 6
TCP/IP is a routable protocol stack that can be run on a number of
different software platforms (Windows, UNIX, and so on) and it is
embraced by most network operating systems as the default network
protocol. TCP/IP contains a number of “member” protocols that
make up the actual TCP/IP stack. And because the TCP/IP protocol
stack was developed before the completion of the OSI reference
model, these protocols do not map perfectly to the various layers of
the model. Figure 2.7 shows the TCP/IP stack mapped to the OSI
layers (the figure provides a general overview of TCP/IP and is not

an exhaustive list of all the protocols in the stack). Table 2.3
describes the protocols listed in the figure. More information will be
provided on all the protocols in the TCP/IP stack in Chapter 10,
“TCP/IP Primer.”
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Mo del and Network Protocols
FIGURE 2.7
TCP/IP is a large proto-
col stack using a number
of member protocols at
various layers of the OSI
model.
4 7
PART I
R e a l - W orld Network Protocols CHAPTER 2
Table 2.3 TCP/IP Protocol Stack Members
Protocol Role
FTP File Transfer Protocol provides an interface and services for
file transfer on the network.
SMTP The Simple Mail Transport Protocol provides email services
on the Internet and IP networks.
TCP The Transport Control Protocol is a connection-oriented
transport protocol. TCP handles a connection
between sending and receiving computers much
like a phone conversation.
UDP User Datagram Protocol is a connectionless transport proto-
col that provides transport services in conjunction with TCP.
IP The Internet Protocol is the basis for all addressing on
TCP/IP networks and it provides a connectionless oriented
Network layer protocol. Works much like an addressed letter

that is dropped in a mail box and then delivered to the
intended destination.
ARP Address Resolution Protocol maps IP addresses to MAC hard-
ware addresses. ARP will be discussed in greater detail in
Chapter 10.
TCP/IP not only provides a very rich set of network-related features
(which means that TCP/IP requires a fair amount of overhead to
run) but also provides a unique logical addressing system. Anyone
connected to the Internet is familiar with the 32-bit IP address,
which is commonly written as 4 octets (an octet being 8 bits of infor-
mation). The typical IP address is written in the format 129.30.20.4,
where each of the four dotted decimal values actually represent 8 bits
of binary information. Much more information concerning IP
addressing will be discussed in Chapter 10.
Because of TCP/IP’s importance in Internetworks and the complexi-
ties related to routing TCP/IP networks, an entire chapter of this
book has been provided reviewing all the aspects of TCP/IP
addressing. A great deal of information will also be provided on the
commands related to routing TCP/IP on a campus or enterprise net-
work.
SEE ALSO
➤ The best place to start in on TCP/IP and routing is Chapter 10,“TCP/IP Primer,” beginning
on page 167.
The IEEE 802
specifications
The IEEE 802specifications
provide categories that
define the Logical Link
Layer and the different net-
work architectures that can

be embraced by the MAC
layer. A complete list of the
802 categories is provided:
• 802.1 Internetworking
• 802.2 Logical Link
Control
• 802.3
Ethernet(CSMA/CD)
LAN
• 802.4 Token Bus LAN
• 802.5 Token Ring LAN
• 802.6 Metropolitan
Area Network
• 802.7 Broadband
Technical Advisory
Group
• 802.8 Fiber Optic
Technical Advisory
Group
• 802.9 Integrated Voice
and Data Networks
• 802.10 Network
Security
• 802.11 Wireless
Networks
• 802.12 Demand
Priority LAN
4 8
IPX/SPX
IPX/SPX (Internetwork Packet Exchange/Sequenced Packet

Exchange) is a network protocol stack developed by Novell for use in
the Novell NetWare network operating system. IPX/SPX is a leaner
stack than TCP/IP and does not require the overhead needed by
TCP/IP. IPX/SPX is suitable for small and large networks and is a
routable network protocol suite.
Figure 2.8 maps protocols in the IPX/SPX stack to the OSI Layers.
Table 2.4 gives a brief description of each of the protocols.
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Model and Network Protocols
TCP/IP trivia
TCP/IP was developed by
Defense Advanced
Research Projects Agency
(DARPA). The Department
of Defense needed a proto-
col stack that could com-
municate across unlike
networks. The unlike net-
works existed because the
government uses a bidding
system and suddenly found
itself with different com-
puter systems at various
branches of the Defense
Department: the Army,
Navy, and so on. So, TCP/IP
is jokingly called the proto-
col of low bid because it
was in part developed to
fix a problem that arose

because of the way the
government takes bids for
procuringtechnology and
other goods.
FIGURE 2.8
IPX/SPX is an efficient
network protocol stack
used on large and small
networks.
4 9
PART I
R e a l - W orld Ne twork Proto cols CHAPTER 2
Table 2.4 IPX/SPX Protocol Stack Members
Protocol Role
SAP The Service Advertising Protocol is used by NetWare File Servers
and Print Servers to announce the address of the server.
NCP The NetWare Core Protocol handles network functions at the
Application, Presentation, and Session layers. It handles packet cre-
ation and is responsible for providing connection services between
clients and servers.
SPX Sequenced Packet Exchange Protocol is a connection-oriented trans-
port protocol
IPX Internetwork Packet Exchange Protocol is a connectionless transport
protocol that handles addressing and routing on the network.
Our major concern with IPX/SPX is routing this protocol suite on
an Internetwork. More information on routing IPX/SPX and how
the IPX/SPX stack moves data on the network is provided later in
this book.
SEE ALSO
➤ Routing IPX/SPX is discussed in Chapter 12,“Routing Novell IPX,” which begins on

page 211.
AppleTalk
While many network administrators would not consider AppleTalk
an Internetworking or enterprise network protocol, AppleTalk is
routable. And with the appropriate type of NIC (Apple Macintoshes
can participate on an Ethernet network if they are outfitted with
EtherTalk cards or other adapters) it can support Ethernet, Token
Ring, and FDDI architectures. It is not uncommon to have
Macintosh computers in the Enterprise to support graphic manipula-
tion and other multimedia duties and so it makes sense to include
AppleTalk as another key routable protocol stack on the corporate
network.
Earlier, in Chapter 1, we discussed AppleTalk as architecture, but it
is also a network protocol stack. Figure 2.9 maps the protocols in the
AppleTalk stack to the layers of the OSI model. Table 2.5 gives a
brief description of each protocol.
Figure alert!
Figures 2.7 through 2.9
map real-world protocols to
the OSI model. To under-
stand these figures, think
back to how the OSI model
describes in seven layers
how data moves from one
computer to another and
the transformation that it
must undergo. Real-world
stacks likeTCP/IP perform
all the tasks described in
the OSI model; they just do

it with fewer protocols.
Rather than having seven
protocols (one for each of
the OSI layers) TCP/IP has
certain protocols that han-
dle the duties of more than
one OSI layer. For example,
FTP handles Application,
Presentation and Session
layer duties. The circle
around FTP spans all three
of the layers on the OSI
model (the layers are the
boxes).
5 0
Table 2.5 AppleTalk Protocol Stack Members
Protocol Role
AppleShare AppleShare provides services at the Application layer
AFP The AppleTalk Filing Protocol provides and managing file
sharing among nodes on the network
ATP The AppleTalk Transaction protocol provides the Transport
layer connection between computers
NBP The Name Binding Protocol maps computer hostnames to
Network layer addresses
ZIP The Zone Information Protocol controls AppleTalk zones and
maps zone names to network addresses
AARP The AppleTalk Address Resolution Protocol maps logical
Network layer addresses to Data Link hardware addresses
PART I Netwo rking Overview
CHAPTER 2 Th e OSI Model and Netwo rk Pro tocols

Terminology alert!
Before we go too much
farther, we should sort out
some terms that you will
find throughout this book:
Internetwork: a network of
networks. Local Area
Networks connected by an
Internetwork device such
as a bridge or router.
Internetworking is dis-
cussed in detail in Chapter
4, “Internetworking
Basics.”
Internet: The globalnet-
work of networks. TCP/IP is
the de-facto standard for
this global collection of
heterogeneous computers.
Intranet: Acorporate net-
work that is internal to the
enterprise (not connected
to the global Internet) but
uses Internet protocols
such as Simple Mail
Transport Protocol and
Hypertext Transport
Protocol (the protocol used
by Web Browsers) to share
information among corpo-

rate users. An extranet is
an intranet that provides
corporate network access
to specified users outside
the company.
FIGURE 2.9
AppleTalk is a routable
protocol stack for
Macintosh networks that
can communicate with
Ethernet, Token Ring,
and FDDI networks.
5 1
PART I
R e a l - W orld Ne twork P rotocols CHAPTER 2
DDP The Datagram Delivery Protocol provides the addressing sys-
tem for the AppleTalk network and provides connectionless
transport of datagrams between computers
As with IPX/SPX, our interest in the AppleTalk network protocol
stack relates to routing AppleTalk. More information on how
AppleTalk networks are configured and how AppleTalk is routed on
a Cisco router is provided later in this book (see Chapter 13,
“Routing Apple Talk.”)
SEE ALSO
➤ More information on how AppleTalk networks are configured and how AppleTalk is routed on
a Cisco router is provided on page 227.
Protocol Role
Where are the routing
protocols?
You might have noticed

that the diagrams that map
various protocol stacks to
the OSI model did not
include routing protocols.
Obviously, each protocol
stack has a default routing
protocol; for example, RIP
is the default routing
protocol for TCP/IP and the
Routing Table Maintenance
Protocol is the routing
protocol for the AppleTalk
stack. These protocols will
be discussed in greater
detail when routing of
these protocol stacks is
discussed later in this
book.

Wide Area Networking
Understanding Wide Area Connectivity
•
Getting Connected
•
Switched Network Overview
•
Circuit Switching
•
Packet Switching Protocols
•

Other WAN Protocols
•
3
c h a p t e r
5 4
Understanding Wide Area Connectivity
As the PC local area network became more and more important to
businesses, corporations, and institutions, the need to expand and
then connect LANs became a necessity. Expanding or connecting
LANs locally (in a fairly limited geographic area) was taken care of
by internetworking devices such as repeaters, bridges, switches, and
routers. However, when connecting LANs over large distances, other
technology must come into play.
A need for technology that provided network administrators with the
ability to connect LANs over greater geographic areas became
extremely important as networking the enterprise (the enterprise is
the entire corporation—which in many cases can be a worldwide
operation) became an imperative.
Expanding a network across great distances can be accomplished by
taking advantage of several different wide area networking technolo-
gies. Networks can be connected with services provided by the pub-
lic switched telephone network (PSTN) or private carrier companies.
Extremely large companies can invest in their own WAN infrastruc-
ture and invest in microwave and satellite transmission equipment.
WAN technology can be used to connect networks between two
cities, across the country, or around the world. As with LANs and
internetworks, after the Physical layer aspects have been taken care
of, various protocols are used to move the data on the WAN. On a
LAN, the cable and the hubs provide the Physical layer, while on a
WAN, the Physical layer can be a T1 leased line or a satellite dish.

SEE ALSO
➤ For more information on internetworking,see page 67.
Getting Connected
While the actual physical infrastructure (the cabling and networking
devices such as hubs, repeaters, and so on) of a LAN will be owned
by a company, most businesses and institutions find it too costly to
own the physical WAN connections that they use.
PART I Netwo rking Overview
CHAPTER 3 Wi de Area Networking
WANs take advantage of
wireless technologies
Although LANs typically
use some sort of physical
wiring (copper or fiber-optic
cable), WANs can take
advantage of several wire-
less technologies, including
microwave transmissions,
satellite links, infrared
light, and network commu-
nications via radio signals
(both single-frequency and
spread-spectrumradio
transmissions).
5 5
PART I
Getting Connected CHAPTER 3
Three types of WAN connections are available: a connection over
the Public Switched Telephone System via a modem, a dedicated
connection such as a full-time leased line, or a switched connection

that enables multiple users to take advantage of the same line.
Each of these WAN connection possibilities offers its own set of pros
and cons and each embraces different hardware needs. The following
sections discuss these three WAN connection alternatives.
Dial-Up Connections
The simplest and least expensive type of dial-up connection uses a
modem to connect two computers over a regular analog voice-grade
telephone line. The modem converts the digital information on the
computer to an analog signal (modulation) and vice versa (demodula-
tion). That’s how the modem got its name. This conversion process
allows computer data to be sent over the analog line. Modems are
now available that have potential transmission speeds of up to
56Kbps, but line noise can limit the speed at which a connection
over an analog line can run.
Routers can be outfitted with a modem connection (and then are
often referred to access servers). This means that two LANs could be
connected via a dial-up connection and packets routed (although this
would provide a very slow connection between the networks). Figure
3.1 shows two LANs connected via a dial-up connection with routers
serving as the connection point for the asynchronous connection
over modems.
Keep the packets on the
company-owned wire
One of the secrets of being
a highly successful WAN or
internetwork administrator
is designing large networks
and setting up the potential
routing of packets so that
traffic circulates on the

company-owned network-
ing infrastructure as much
as possible. Running a
cost-effective internetwork
or WAN is the ultimate
challenge when involving
leased lines and connec-
tions for which you basi -
cally pay for the
bandwidth.
FIGURE 3.1
LANs can be connected
using dial-up connec-
tions via modems.